Computer scientists at the University of Nottingham along with Kingston University have resolved a complex problem in the field of vision and graphics research. They have developed a technology that holds the potential of mapping 3D facial reconstruction from a single 2D image i.e., a 3D selfie.
By just uploading and receiving a single color image, a 3D model showing the shape of their face. Individuals are lining up to attempt it thus far, more than 400,000 users have had a go.
Scientists developed this technique by using a Convolutional Neural Network (CNN). Convolutional Neural Network (CNN) is a part of artificial intelligence (AI), where machine learning algorithms, give computers the ability to learn without being explicitly programmed. However, the technique is just a prototype.
Here, scientists trained CNN on a huge dataset of 2D pictures and 3D facial models. Thus, CNN potentially reconstructed 3D facial geometry from a single 2D image. Along with this, it also identified the non-visible parts of the face.
Dr. Yorgos Tzimiropoulos said, “The main novelty is in the simplicity of our approach which bypasses the complex pipelines typically used by other techniques. We instead came up with the idea of training a big neural network on 80,000 faces to directly learn to output the 3D facial geometry from a single 2D image.”
Aaron Jackson said, “Our CNN uses just a single 2D facial image, and works for arbitrary facial poses (e.g. front or profile images) and facial expressions (e.g. smiling).”
“The method can be used to reconstruct the whole 3D facial geometry including the non-visible parts of the face.”
Furthermore, this technique of creating 3D selfie has more standard applications. This technology could be used to personalize computer games, improve augmented reality, and let people try on online accessories such as glasses.
Scientists suggest, “It could also have medical applications too. Using it in surgeries like plastic surgeries could bring more simulating results.”